APPLICANT'S ABSTRACT: The overall goal of this project is to conduct long range follow-up studies of two cohorts of students who participated in school-based adolescent alcohol misuse prevention studies in the elementary and junior high school grades, in order to test the long-range efficacy of the prevention approach and to further test the validity of theoretical models of antecedents of adolescent alcohol use and misuse. The first cohort (AMPS1) consists of about 1,700 students who participated in a pretest/no test, treatment/control group longitudinal study beginning in grades 5 and 6 in 1984-85. These students are now in grade 12 or one year post-high school. The second cohort (AMPS2) participated in an enhanced intervention study, with the intervention taking place in their grades 6, 7, and 8 during 1989-1992. These (approximately 3,000) students are now finishing grade 8. The specific aims of the proposed project are to:1) track and collect data on alcohol use and misuse and the primary intervening variables from the original AMPS1 subjects on three years post-high school) in a mailed questionnaire and telephone interview; 2) track and collect data during the early grade 11 school year, by classroom administered questionnaires for those still in the AMPS2 study schools, and by mailed questionnaires and telephone interviews for those who are not, on all study variables from the AMPS2 subjects; 3) conduct in-person interviews with parents of AMPS2 students living at home to provide an additional data source concerning the family variables included in the theoretical model; 4) conduct data analyses for both cohorts separately to determine the long-range effectiveness of the school-based refusal/social skills intervention by comparing treatment and control groups, and by stratifying according to experience with alcohol use prior to the intervention; 5) conduct longitudinal structural equation model tests to assess and compare the long-term stability of the constructs and predictive validity of the theoretical model for both cohorts, through three years post-high school for the first cohort and through early grade 11 for the second cohort, that includes intrapersonal, peer interpersonal, peer alcohol use, and parental alcohol use for the both cohorts, and family latent variables (e.g., such as norms regarding alcohol use, parental monitoring, nurturance, permissiveness, and sibling alcohol use) for the second cohort, and 6) test the intervention effectiveness and theoretical models separately for each gender. Design effect corrections will be applied to take into account the degree of intraclass correlation on the dependent variables.